This is an application for a Collaborative R01 (involving two independent sites: Tufts-New England Medical Center; PI: Susan Folstein; and, The University of Iowa; PI: Joseph Piven) to continue our efforts to find genetic loci linked to the autism phenotype. Following an administrative supplement in year 2 of this 3 year grant we were funded to: 1) complete ascertainment, diagnostic evaluation and blood collection on 150 autism sib pair families; and, 2) to genotype the first screening set of 75 sib pair families. We have surpassed the goals of this project: in the first 2 years we have completed a genome screen (349 markers) of 76 sib pair families (Stage 1 Screening Set). By the end of Year 3 we will ascertained, collected DNA and completely assessed 176 autism sib pair families; collected DNA and completely assessed 80 proband-parent trios; and, ascertained 60 sized pedigrees for a genome screen proposed for the next 5 year granting period, and done some follow-up work on findings from the Stage 1 screen. In our Stage 1 screen we have identified 22 regions providing suggestive evidence of linkage for follow up in our Stage 2 screen using an additional 100-130 families. Additionally, we have strong evidence of linkage to a locus on chromosome 13 (D13S800) with a maximum multi-point heterogeneity lod score of 3.7. 35-40% of our families appears to be linked to this locus. We have found maximum lod scores > 2.0 on chromosomes 4 and 16, and suggestive evidence of linkage to a region on chromosome 7 which has been recently reported to show linkage in the only other genome screen of autism published to date. Finally, we have detected several possible duplications and deletions in the q11-13 region of Chr 15, confirming other reports. Based on these Preliminary data, it appears likely that locus heterogeneity is a factor in the genetic etiology of autism. In the next five year granting period we propose to employ a variety of novel and complementary strategies (including varying phenotypic definition, family structures examined, and molecular and analytic approaches) available through the rich and varied expertise of our collaborative group of senior clinical, molecular and statistical genetics investigators, to resolve the genetic complexity of this disorder. Specifically we propose to: 1) follow up the Stage 1 findings in 100-130 autistic sib pairs families; 2) repeat the two-stage analysis of the nuclear (sib pair) families, adding in information on the Pervasive Developmental Disorder (PDD) and broader autism (BAP) phenotypes in first-degree relatives without autism; 3) conduct a linkage study of autism, as well as a study of the PDD/BAP phenotype, in 60 moderate size pedigrees; and, 4) examine the prevalence of chr. 15 duplications/deletions in autism, and at a large group of controls.